Distortion from audio power amp

[Glenn Gundlach]

Terry Pinnell <terrypinDELETE@THESEdial.pipex.com> wrote in message news:<trn1b0d8th6uslq98btss8stcvv9dovk8r@4ax.com>...

Terry Pinnell <terrypinDELETE@THESEdial.pipex.com> wrote:

legg@magma.ca (R.Legg) wrote:

Terry Pinnell <terrypinDELETE@THESEdial.pipex.com> wrote in message news:<2v9ua0h8fu8d5mh8bhjhnq4eooim544i0t@4ax.com>...
Here's the output waveform from circuit as it stood yesterday
http://www.terrypin.dial.pipex.com/Images/PushPullOutMay22-1.gif

I have yet to test the very latest circuit, incorporating a couple of
small changes:
http://www.terrypin.dial.pipex.com/Images/PushPull-May22-1.gif

Und zo - ze qveschun: vy ist ze zmaller vafeform more distorted zan ze
bigger vafeform?

Diz iz amblivyer, yah?

Id has negadiv distorzhun?

Such genyus!

Good point - hadn't noticed that. Signal is from my Philips PM5134
function generator. To rule out its being some artifact of my PC-based
PicoScope, I'll view with my analog 'scope too.

As I suspected, that was merely a characteristic of my ADC 'scope at
that particular scale.

I've just finished getting the circuit back onto stripboard and it
looks fine. About 7.3 V rms into 8 ohms.
http://www.terrypin.dial.pipex.com/Images/PushPullOutMay23.gif

On my Hameg analog 'scope, both input and output look even cleaner of
course.

Well you're certainly persistent but its a hell of a lot of work for
almost 7 watts. It will be interesting to see it it ducks under the 1%
point on distortion. I have doubts. I must say it has been a LONG time
since I've seen output coupling caps and center voltage tweaks.
GG

 

[Terry Pinnell]

stratus46@yahoo.com (Glenn Gundlach) wrote:

Well you're certainly persistent but its a hell of a lot of work for
almost 7 watts. It will be interesting to see it it ducks under the 1%
point on distortion. I have doubts. I must say it has been a LONG time
since I've seen output coupling caps and center voltage tweaks.

And I'm not sure I'll be able to answer that question. So far I've had
no responses at all to my questions in
Newsgroups: sci.electronics.design
Subject: Re: Distortion from audio power amp
Date: Fri, 21 May 2004 16:11:05 +0100
Message-ID: <ke6sa0h3n1phi2542ddepv5jb1mbepojrd@4ax.com>

Yes, it's very much an issue of curiosity and intellectual
satisfaction now! And if I confine the amp to its original purpose,
delivering a pseudo-random, simulated, burglar-deterrent dog bark, as
described up-thread, then even 1% THD is arguably well OTT <g>.


--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Stefan Heinzmann]

Terry Pinnell wrote:

stratus46@yahoo.com (Glenn Gundlach) wrote:

Well you're certainly persistent but its a hell of a lot of work for
almost 7 watts. It will be interesting to see it it ducks under the 1%
point on distortion. I have doubts. I must say it has been a LONG time
since I've seen output coupling caps and center voltage tweaks.


And I'm not sure I'll be able to answer that question. So far I've had
no responses at all to my questions in
Newsgroups: sci.electronics.design
Subject: Re: Distortion from audio power amp
Date: Fri, 21 May 2004 16:11:05 +0100
Message-ID: <ke6sa0h3n1phi2542ddepv5jb1mbepojrd@4ax.com

Yes, it's very much an issue of curiosity and intellectual
satisfaction now! And if I confine the amp to its original purpose,
delivering a pseudo-random, simulated, burglar-deterrent dog bark, as
described up-thread, then even 1% THD is arguably well OTT <g>.

A distorted dog-bark may sound even more vicious than a high-fidelity
one

I can understand your curiosity. This amplifier design is in its
principles almost 50 years old. So the performance will be typical of
designs of that vintage. It would be interesting to gradually apply
improvements that have been invented in the meantime, so apart from
getting a better amp you also follow the trail of amplifier development
in the past 50 years, educating yourself in the process. It'll keep you
busy for a while, anyway...

To give at least /one/ answer to the questions you posted earlier:

Lacking a distortion meter, I'm looking for ways of roughly assessing
the total harmonic distortion in my output signal. For example, at
what point does harmonic distortion become visible? Assume say a
screenshot captured from my PC-based 'scope and displayed on a typical
17" PC screen. I presumably would not see 0.1%, and I definitely would
see 10% - but where is the threshold?

Generalising from that, anyone know a source of illustrations of a
range of values of distortion in a 1 kHz sine wave? I could then
possibly make my estimate by visual comparison.

Alternatively, my 'scope can also show a Fourier Analysis. Would I get
a reasonable estimate if I went through the chore of identifying the
significant components, converting them to voltages, squaring them,
adding them, and taking the square root?

Are there any other relatively methods anyone can recommend please?

You can not usually see distortion of about 1% or less on a scope. You
may have more luck with the Fourier analysis. You should be able to see
harmonics if they are prominent enough. It depends on the specifications
of your scope how far down it shows any useful result. And the result is
not giving you a neat percentage figure, but rather should show the
nature of the distortion, which is of course helpful for figuring out
where it originates.

If you haven't got a distortion tester, you could try to help yourself
with a low distortion sine wave generator, a notch filter and a true RMS
millivolt meter. The process is somewhat laborious, as you have to tweak
the notch filter until you manage to remove the base frequency, leaving
the distortion to be measured. Watching the resulting signal (with the
base frequency removed) on the scope is also quite revealing.

The measurement accuracy is of course critically dependent on the
quality of the oscillator and the notch filter. I don't know what you've
got, but a scope alone won't get you very far.

If your amp had a differential input stage, which would be a good idea
anyway, you could also try to measure the difference between the two
inputs of the long tailed pair, but that won't directly give a
distortion figure either.

--
Cheers
Stefan

 

[Terry Pinnell]

Stefan Heinzmann <stefan_heinzmann@yahoo.com> wrote:

A distorted dog-bark may sound even more vicious than a high-fidelity
one

I can understand your curiosity. This amplifier design is in its
principles almost 50 years old. So the performance will be typical of
designs of that vintage. It would be interesting to gradually apply
improvements that have been invented in the meantime, so apart from
getting a better amp you also follow the trail of amplifier development
in the past 50 years, educating yourself in the process. It'll keep you
busy for a while, anyway...

To give at least /one/ answer to the questions you posted earlier:

Lacking a distortion meter, I'm looking for ways of roughly assessing
the total harmonic distortion in my output signal. For example, at
what point does harmonic distortion become visible? Assume say a
screenshot captured from my PC-based 'scope and displayed on a typical
17" PC screen. I presumably would not see 0.1%, and I definitely would
see 10% - but where is the threshold?

Generalising from that, anyone know a source of illustrations of a
range of values of distortion in a 1 kHz sine wave? I could then
possibly make my estimate by visual comparison.

Alternatively, my 'scope can also show a Fourier Analysis. Would I get
a reasonable estimate if I went through the chore of identifying the
significant components, converting them to voltages, squaring them,
adding them, and taking the square root?

Are there any other relatively methods anyone can recommend please?

You can not usually see distortion of about 1% or less on a scope. You
may have more luck with the Fourier analysis. You should be able to see
harmonics if they are prominent enough. It depends on the specifications
of your scope how far down it shows any useful result. And the result is
not giving you a neat percentage figure, but rather should show the
nature of the distortion, which is of course helpful for figuring out
where it originates.

If you haven't got a distortion tester, you could try to help yourself
with a low distortion sine wave generator, a notch filter and a true RMS
millivolt meter. The process is somewhat laborious, as you have to tweak
the notch filter until you manage to remove the base frequency, leaving
the distortion to be measured. Watching the resulting signal (with the
base frequency removed) on the scope is also quite revealing.

The measurement accuracy is of course critically dependent on the
quality of the oscillator and the notch filter. I don't know what you've
got, but a scope alone won't get you very far.

If your amp had a differential input stage, which would be a good idea
anyway, you could also try to measure the difference between the two
inputs of the long tailed pair, but that won't directly give a
distortion figure either.

Thank, that's good stuff.

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Jim Thompson]

On Mon, 24 May 2004 17:32:29 +0100, Terry Pinnell
<terrypinDELETE@THESEdial.pipex.com> wrote:

[snip]

Lacking a distortion meter, I'm looking for ways of roughly assessing
the total harmonic distortion in my output signal. For example, at
what point does harmonic distortion become visible? Assume say a
screenshot captured from my PC-based 'scope and displayed on a typical
17" PC screen. I presumably would not see 0.1%, and I definitely would
see 10% - but where is the threshold?

[snip]

Play a Mozart woodwind ensemble (with French Horn) thru it. If you
have significant distortion you'll plainly hear it as an _awful_
dissonance.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

 

[Terry Pinnell]

"petrus bitbyter" <p.kralt@reducespamforchello.nl> wrote:

Terry,

IMHO R11 is at the rihgt place, but the emiiter of Q4 has to be connected to
the collector of Q6. Found this and some other tips in an (old? 1973?) book,
on the subject. This type of amplifiers were never known for their
stability. I posted a schematic on alt.binaries.schematics.electronic that
contains eight clues to stabilise an amplifier like this. I consider the
author a specialist as I did a lot of amplifier (re)building using this book
and never failed.

I saw the schematic, thanks, here:
Newsgroups: alt.binaries.schematics.electronic
Subject: Hobbyists old amplifier
Message-ID: <Hevpc.220048$s9.153453@amsnews02.chello.com>
Date: Sat, 15 May 2004 20:40:07 GMT

But was there also any explanatory text about those 'eight clues'
please?

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[Terry Pinnell]

John Popelish <jpopelish@rica.net> wrote:

Terry Pinnell wrote:
(snip)
I intend to replace the two presets for setting the output DC level
and quiescent current respectively. The first is obviously easy to set
(half supply voltage), but how do I determine optimum setting for
quiescent current please?

The voltage between the bases of Q3 and Q4 have to not only force
those two transistors into slight conduction, but Q5 and Q6, also. I
would connect a millivolt meter across R11 and raise the bias voltage
until a few millivolts drop was visible. But, as I said in an earlier
post, this will require at least 3 diode drops between the basses of
Q3 and Q4. After you set the quiescent current through R11, go back
and measure one of the diode drops (D1 or D2) and the total voltage
between the bases of Q3 and Q4 to see if I am right about this.

Thanks. Across R11 I get 1mV immediately after power up, rising to
around 3mV after 5 to 10 minutes, implying 15mA quiescent current.
Seems pretty steady at that level. That's with my R7 combination of 15
+ 150 preset at about 158R, close to its max, surprisingly far above
the 68R Torrens specified for a 24V supply. With a signal applied, I
was getting 7.3V rms unclipped.

When measuring the Q3B-Q4B voltage difference, I first placed my DMM
directly across those nodes, and got violent oscillation. Then I
realised that was hardly surprising, as I was removing the diode
action. So I measured each point independently. But although Q3B was
what I expected from simulation, 13.1V, Q4B was wildly different, at
630mV!

Here's a table showing simulated results. It reinforces your point
about 3 junction drops for most of the range shown:

Ohms ma .............Volts...........
R7 R11 Q3B Q4B Diff Q1B DC Out
---- ---- ---- ---- ---- ---- ------
22 2.1 13.0 11.2 1.8 1.53 12.0
33 2.4 13.0 11.2 1.8 1.53 12.0
47 2.8 13.1 11.2 1.9 1.53 12.0
68 2.1 13.1 11.2 1.8 1.53 12.0 Orig. recommended value
100 4.6 13.2 11.2 2.0 1.53 12.0
120 7.2 13.3 11.1 2.2 1.53 12.0
130 11.7 13.3 11.1 2.2 1.53 12.0
140 22.1 13.4 11.1 2.3 1.53 12.0
150 41.5 13.4 11.1 2.3 1.53 12.0
158 63.9 13.4 11.1 2.3 1.53 12.0
158 3.0 13.1 0.63 12.5 1.18 12.0 Actual result
---------------------------------------------------------
200 240 13.5 11.1 2.4 1.53 12.0
1k 3000 14.3 10.4 3.9 1.52 11.9

--
Terry Pinnell
Hobbyist, West Sussex, UK

 

[John Popelish]

Terry Pinnell wrote:

Thanks. Across R11 I get 1mV immediately after power up, rising to
around 3mV after 5 to 10 minutes, implying 15mA quiescent current.
Seems pretty steady at that level.

That implies something like 24*.015=.36watts of power lost to reduce
the distortion relative to Torrens design. This is pretty reasonable,
compared to the 6 watts delivered to the speaker.

That's with my R7 combination of 15
+ 150 preset at about 158R, close to its max, surprisingly far above
the 68R Torrens specified for a 24V supply. With a signal applied, I
was getting 7.3V rms unclipped.

His design isn't concerned about cross over distortion as much as
power dissipation and thermal stability, evidently.

When measuring the Q3B-Q4B voltage difference, I first placed my DMM
directly across those nodes, and got violent oscillation. Then I
realised that was hardly surprising, as I was removing the diode
action.

I don't understand this. A high impedance volt meter (10 meg ohm
typical) should not have been much of a shunt in parallel with the
diode resistor bias combination. It will add quite a bit of stray
capacitance.

So I measured each point independently. But although Q3B was
what I expected from simulation, 13.1V, Q4B was wildly different, at
630mV!

To the negative supply rail? Can't be right.

Here's a table showing simulated results. It reinforces your point
about 3 junction drops for most of the range shown:

Ohms ma .............Volts...........
R7 R11 Q3B Q4B Diff Q1B DC Out
---- ---- ---- ---- ---- ---- ------
22 2.1 13.0 11.2 1.8 1.53 12.0
33 2.4 13.0 11.2 1.8 1.53 12.0
47 2.8 13.1 11.2 1.9 1.53 12.0
68 2.1 13.1 11.2 1.8 1.53 12.0 Orig. recommended value
100 4.6 13.2 11.2 2.0 1.53 12.0
120 7.2 13.3 11.1 2.2 1.53 12.0
130 11.7 13.3 11.1 2.2 1.53 12.0
140 22.1 13.4 11.1 2.3 1.53 12.0
150 41.5 13.4 11.1 2.3 1.53 12.0
158 63.9 13.4 11.1 2.3 1.53 12.0
158 3.0 13.1 0.63 12.5 1.18 12.0 Actual result

Don't believe this last one. Something is wrong with the way you
measured it.

---------------------------------------------------------
200 240 13.5 11.1 2.4 1.53 12.0
1k 3000 14.3 10.4 3.9 1.52 11.9

The last one here should have overloaded your power supply.

--
John Popelish

 

[Stefan Heinzmann]

Glenn Gundlach wrote:
[...]

Many DIY magazine projects in Popular Electronics and
Radio-Electronics RCA transistor manual were diff in, dual NPN
outputs.

Speaking of the RCA transistor manual, does anyone know whether there's
a scanned version of it available for download somewhere?

--
Cheers
Stefan